Artificial Intelligence (AI) technology has already made its way into our lives, in travel, consumption, entertainment, and research. You may not have noticed, but AI technology has also made its way into the medical field, where it can help doctors make quick decisions and even double the speed of new drug development. Wang Congjun, president of Beijing Tiantan Hospital of Capital Medical University, detailed the application of AI technology in the clinic .
The origins of AI and medicine
In 1956, John McCarthy, the founder and CEO of Dartmouth College, introduced AI to the medical community. In 1956, at Dartmouth College, John McCarth, Marvin LeeMinsky, and others were invited to present their work on AI. Marvin LeeMinsky, Claude Elwood, and John McCarthy at Dartmouth College in the United States. Claude Elwood Shannon. Claude Elwood Shannon, Nathaniel R. Rochester, and Nathaniel R. Rochester. Nathaniel Rocheste (Nathaniel Rocheste) four scholars initiated a summer symposium on AI, pointed out that the "AI" research goal is to realize a machine that can simulate human beings, the machine can use language, have the ability to understand, can complete the tasks that only human beings can do, and constantly improve the machine itself. The concept has been in development for 65 years now.
"Seeing the definition of AI it is easy to see that AI is closely linked to brain science and clinical neuroscience, and AI networks are designed to mimic the human brain, with the ultimate AI being the realization of brain-like computing." Wang Congjun said.
When AI first appeared, it was already being used in medicine, first in the form of electronic medical records. By 2010 and beyond, AI was being used to process genomic big data and enable natural language recognition on electronic medical records.
Wang Congjun introduced that from 2019, the application of AI technology in the field of medicine went to the next level, used to find new drug targets and drug development.
In 2020, the journal Nature Medicine published two important guidelines SPIRIT-AI EXTENSION and CONSORT-AI EXTENSION, which give AI a basis for entering clinical applications, and are two important programmatic documents to promote the application of AI in the clinic.
AI makes stroke patients no longer subject to the limitations of the golden period for resuscitation
AI is a good helper for doctors, because its appearance has changed the rescue mode of stroke patients.
As we all know, when a stroke occurs, it needs to be implemented in the golden time for rescue - intravenous thrombolysis within 4.5 hours and mechanical thrombolysis within 6 hours, and once it exceeds this time, it is very easy to be disabled or even die.
Wang Qunjun said: "After the stroke, the middle of the brain artery trunk occlusion in the middle of a core necrotic area, outside there is a circle of ischemia, but not necrotic, this time is the window of the doctor's rescue. Under normal circumstances, every 100 grams of brain tissue through the blood flow of about 100 milliliters per minute, when down to 20 milliliters, sodium and potassium ions in the cell membrane of the internal and external exchanges will stop, less than 10 milliliters when the cell membrane began to rupture, we will be 10 to 20 milliliters of blood flow period is called the semi-dark zone, for the doctor to provide a window of salvation."
Back in the 1980s, recognizing the semi-dark band was a challenge, and without intelligent means, it could only be recognized by the doctor looking at the film, which affected the speed of the emergency. With AI technology, this traditional approach has been revolutionized, and not only are semi-dark bands better identified, but there is no need to consider the resuscitation time window.
Drug R&D from knowledge-driven to data-driven
Drug R&D is a very high coefficient of difficulty of scientific research, each new drug from the beginning of the research and development to the market application will have to go through more than ten years or even decades of the course, and the emergence of AI accelerates the process.
AI has accelerated this process.
"Before 2015, our drug R&D model was the traditional model, and the general process was to identify disease targets and then conduct experiments on animal cell models, organ models, and animal models, and then conduct clinical trials, a process that took more than a decade.After 2015, AI began to be applied to drug discovery and development, disrupting the entire R&D process, where we interacted imaging data, histology data, clinical data, and environmental data together to form big data, and used AI to find mathematical relationships between these data, which in turn led to the creation of innovative drugs and innovative reagents."
This shift in process is also indicative of a shift in the drug development model from knowledge-driven to data-driven. Wang Congjun said, "The reverse drug development model of clinical big data and multi-omics, referred to as the BBB model, combines knowledge-driven and data-driven, and the performance of the finished drug is improved by 100 times, and the research and development time will be shortened by half, and this is the new change that AI brings us."
In addition to this, AI is also being used in many aspects of the medical field, such as the use of the da Vinci surgical robot, and in Japan there are also AI companion devices, room-checking robots, and so on, which work by machine to improve efficiency and reduce the doctors' potential hazards.
Wang Congjun said, "One of the reasons AI is growing so quickly in clinical neuroscience is that we've made AI a partnership that allows clinicians to use AI technology earlier and smoother, which also brings benefits to patients."
Typeset by Tao Tao